Evaluating the parental fitness of psychiatrically diagnosed individuals: advocating a functional-contextual analysis of parenting.

Abstract

The parental fitness of psychiatrically diagnosed individuals is often questioned in termination of parental rights cases. The goal of this article is to shift the focus from a predisposing bias of unfitness to a functional-contextual analysis of parenting behavior and competency. Three underlying biased assumptions are relevant for the courts' decision making: (a) that a diagnosis (past or present) predicts inadequate parenting and child risk, (b) that a diagnosis predicts unamenability to parenting interventions, and (c) that a diagnosis means the parent is forever unfit. Each assumption will be considered in light of empirical evidence, with major depression, schizophrenia, substance abuse, and mental retardation provided as examples of diagnostic labels often assumed to render a parent unfit. A research agenda to improve clinicians' ability to assess parental fitness and understanding of how parental mental illness, mental retardation, or substance abuse might compromise parenting capacities is discussed for forensic purposes.

Full text

Page 1

Evaluating the Parental Fitness of Psychiatrically Diagnosed
Individuals: Advocating a Functional–Contextual Analysis of Parenting
Corina Benjet
National Institute of Psychiatry, Mexico City, Mexico
Sandra T. Azar
Pennsylvania State University
Regina Kuersten-Hogan
Clark University
The parental fitness of psychiatrically diagnosed individuals is often questioned in termina-
tion of parental rights cases. The goal of this article is to shift the focus from a predisposing
bias of unfitness to a functional–contextual analysis of parenting behavior and competency.
Three underlying biased assumptions are relevant for the courts’ decision making: (a) that a
diagnosis (past or present) predicts inadequate parenting and child risk, (b) that a diagnosis
predicts unamenability to parenting interventions, and (c) that a diagnosis means the parent
is forever unfit. Each assumption will be considered in light of empirical evidence, with major
depression, schizophrenia, substance abuse, and mental retardation provided as examples of
diagnostic labels often assumed to render a parent unfit. A research agenda to improve
clinicians’ ability to assess parental fitness and understanding of how parental mental illness,
mental retardation, or substance abuse might compromise parenting capacities is discussed
for forensic purposes.
Mentally ill women, like mentally well women, bear
children, and usually share the same desire to raise their
children. However, when women with these diagnoses are
involved in the most extreme of custody proceedings,
namely termination of parental rights hearings, parental
unfitness may be assumed from diagnosis without close
examination of how the disorder specifically impacts on
their parenting. Thus a lower threshold for the termination
of parental rights for these women is set. This lower thresh-
old is reflected in some states’ statutes, which permit ter-
mination of parental rights due to mental illness, mental
deficiency, habitual use of alcohol or drugs, “debauchery,”
or “repeated lewd and lascivious behavior” (Melton, Petrila,
Poythress, & Slobogin, 1997, p. 466). In this article we
consider severely depressed, schizophrenic, substance abus-
ing, and mentally retarded parents as parental groups that
often come under scrutiny in order to illustrate the impact of
a diagnostic label on the assessment of parental fitness.
Though parental pathology does influence parenting quality
and capabilities, so does poverty, parental stress, and pa-
rental physical illness. However, we do not remove children
from their parents solely because their parents are poor,
stressed, or physically ill. As recently as 1980, one state’s
statute allowed for the termination of parental rights of
mentally retarded persons without their consent or even any
judicial determination that they were unfit (Field &
Sanchez, 1999). More recent case law has emphasized that
courts need to make custody judgments on the basis of an
evaluation of parental competency and child risk, not by the
mere presence of mental illness alone. Nevertheless, it has
been argued that much discretion is still allowed in such
cases (Grisso, 1986).
“There are few other areas of law where the courts rely as
heavily on social science data as they do for decisions about
children’s welfare” (G. S. Goodman, Emery, & Haugaard,
1998, p. 775), and as social scientists we have the ethical
responsibility to inform the courts with information that is
empirically based. The purpose of this article, then, is to
challenge the automatic assumption of parental unfitness of
psychiatrically diagnosed parents (whatever that diagnosis
might be) in order to sensitize expert witness evaluations
and testimony and, ultimately, judges’ decision-making. We
urge consideration of factors that might determine whether
and in what ways the parenting of persons with these
diagnoses might be compromised, and we propose a more
functional–contextual analysis of parenting capability. To
do this, we consider the most extreme custody decision,
termination of parental rights, and the most scrutinized
parents, those with a severe affective or psychotic disorder,
substance abuse disorder, or mental retardation. First, we
briefly summarize the legal requirements and the human
decision-making processes that are involved in the termina-
Corina Benjet, Department of Epidemiological and Psychoso-
cial Research, National Institute of Psychiatry, Mexico City, Mex-
ico; Sandra T. Azar, Department of Psychology, Pennsylvania
State University; Regina Kuersten-Hogan, Department of Psychol-
ogy, Clark University.
The writing of this article was supported by a National Institute
of Mental Health (NIMH) FIRST Grant Award (NIMH Grant
MH46940) and the award of a Liberal Arts Fellowship at Harvard
Law School (2000–1) to Sandra T. Azar.
Correspondence concerning this article should be addressed to
Sandra T. Azar, Department of Psychology, Moore Building,
Pennsylvania State University, University Park, Pennsylvania
16802. E-mail: sta10@psu.edu
Journal of Family Psychology
2003, Vol. 17, No. 2, 238–251
Copyright 2003 by the American Psychological Association, Inc.
0893-3200/03/$12.00DOI: 10.1037/0893-3200.17.2.238
238

Page 2

tion of parental rights. Next, we review the research litera-
ture as it pertains to three broad assumptions regarding the
psychiatrically diagnosed parent that might color the expert
witness’s testimony and, ultimately, the judge’s decision.
These assumptions are that (a) a diagnosis (past or present)
predicts inadequate parenting or risk for child maltreatment,
(b) a diagnosis predicts a lack of amenability to parenting
interventions; and (c) a diagnosis means that the parent is
forever unfit. Having reviewed the empirical evidence for
these assumptions, we then outline a research agenda that
might enhance mental health professionals’ ability to pro-
vide balanced information to courts regarding the psychiat-
rically diagnosed parent and the viability of keeping the
family united. Additionally, we suggest domains of the
parenting context specific to evaluating psychiatrically di-
agnosed parents that should be examined.
Termination of Parental Rights
Legal Requirements
The process of terminating parental rights most typically
begins with an initial report of abuse or neglect. (However,
it should be noted that in the case of mentally ill, substance-
abusing, or developmentally disabled parents, a tendency
toward “predictive determinancy” has been suggested,
whereby suspicions begin even before the parent has a
chance to do harm, that is, during pregnancy, which may
lead to removal of children at birth or undue pressure to
voluntarily give up custody of the child or terminate preg-
nancy; Field & Sanchez, 1999). Once reported, the state
investigates and takes emergency measures in the form of
temporary removal of the child if deemed necessary. De-
pending on the conclusions of the investigation, the state
attorney or social worker may then file a petition, following
which disposition is determined. Ideally, custody is tempo-
rarily transferred while parents are required to meet condi-
tions set by the court in order to improve parenting fitness
and regain custody of the child. If the parent is unable to
meet the conditions set forth by the court, a further hearing
is likely to consider termination of parental rights (Melton et
al., 1997). Whereas services are often offered by the state to
help the parent meet these conditions, in the case of parents
with psychiatric diagnoses, this process may be shortened or
services either not offered or offered in a form that may not
meet their special needs.
Mental health providers are often involved in the process
as expert witnesses to inform the court either by providing
a written report or directly testifying in court on the basis of
evaluations of the parent and child, interviews with service
providers working with the family, and reviews of child
protective service records.
Before the state can sever the bond between a parent and
child, the parent must demonstrate “unfitness” that is not
amenable to intervention. The state must also demonstrate
that it provided the services necessary to bring the parent to
a level of fitness such that they can resume parenting and
that these interventions were either not utilized or the parent
did not show adequate progress (Melton et al., 1997).
In the past, the mere presence of certain conditions (such
as parental mental illness, mental retardation, or certain
life-styles, e.g., drunkenness, immorality) was sufficient to
result in termination of parental rights (Grisso, 1986). Re-
cently, however, it has been further required that these
conditions be shown to significantly affect the child’s wel-
fare and that the evidence be “clear and convincing”
(Schetky & Benedek, 1992). The threshold for making
such decisions, however, varies greatly depending on
the jurisdiction’s philosophy and is thus open to much
interpretation.
The criteria for determining fitness have not been well
specified in the law. Generally, there are two considerations
relevant to parental fitness: imminent risk of harm to the
child and minimally adequate care within community stan-
dards. Because legal definitions of parental fitness, abuse,
and neglect are vague and varied, courts have turned to
mental health professionals as expert witnesses to inform
them on this topic with the assumption that a database and
criteria for determining fitness do exist within the mental
health field. This assumption may be less than sound (Azar,
1991). Universal criteria for parenting competency, let
alone well-validated assessment techniques for these crite-
ria, are not yet available.
Human Decision Making
In the absence of clear statutory guidance and profession-
ally validated and agreed upon criteria, human cognitive
processes must be considered in understanding how parent-
ing competency is determined. These processes guide
decision-making and person perceptions and are heavily
influenced by societally determined expectancies, and are
hence, open to the biases inherent in societal views. Cog-
nitive processes favor categorical thinking in the service of
efficiency (Macrae & Bodenhausen, 2000). These general
cognitive processes governing human thought have impli-
cations for both judges and evaluators. Levi (1949), in his
classic discussion of legal reasoning, argued that common
beliefs (i.e., the meaning ascribed to the words put forth in
statutes) shape interpretation of criteria in the law. This
meaning, he argues, can contradict the intent of the
legislature.
Psychological testimony also may be biased by stereo-
types of the mentally ill, which are founded more on societal
expectancies than on empirical evidence. Complicating
mental health evaluations is the fact that the legal judgments
of parental fitness in termination cases are based on criteria
of providing less than minimally adequate care and placing
a child at imminent risk of harm. Mental health profession-
als, however, have a bias toward advocating optimal family
environments for children. Consideration of minimally ad-
equate conditions is not part of the professional schema on
which theories are built and tested. Thus, there may be
discordance in the information that the legal system requires
to make a termination decision and the information mental
health professionals can provide.
Societal Schema Regarding the Mentally Ill Parent
Historically, the mentally ill have been viewed with ex-
treme disfavor and have been treated poorly by the com-
239 SPECIAL ISSUE: PARENTAL FITNESS AND MENTAL ILLNESS

Page 3

munities in which they have lived. Even in modern times,
individuals who have a diagnosis of mental disorder con-
tinue to be more likely to be viewed as incompetent. Re-
search has documented that, presented with an equal stim-
ulus presentation (e.g., a research confederate, written
description), others will judge a person labeled as mentally
ill more harshly than one without this label (Farina, 1982).
The mentally ill are less likely to be seen as strong prospects
for jobs or as tenants (Farina & Felner, 1973; Page, 1977),
and are more likely to receive harsh treatment (e.g., longer
shocks in a learning task; Farina, Holland, & Ring, 1966).
Conversely, if individuals are told that the person they are
about to meet believes they have a mental illness, their
performance is negatively affected, even when that person
has no such information (Farina, Allen, & Saul, 1968). This
suggests that such individuals may behave differently in
situations where their condition is “known” (e.g., in a par-
enting evaluation). Moreover, attitudinal research also has
shown that mentally ill persons are viewed as more perma-
nently at a disadvantage than another stigmatized group
(convicts; Lamy, 1966). Even mental health professionals
are biased in interpreting the behavior of diagnostically
labeled individuals (Langer & Abelson, 1974; Rosenhan,
1973).
Such biases may influence judgments regarding parental
fitness, amenability to intervention, and the level of risk to
others. Indeed, in judging the domains most affected by a
number of disabling conditions, emotional disorders were
viewed as most detrimental to parenting and, in fact, par-
enting was seen as the domain most likely to be affected by
emotional disorder (more than marital relations, vocation,
etc.; MacDonald & Hall, 1969). Legal writers have cited
similar biases in custody cases involving particular diagnos-
tic groups. For example, with regard to mentally retarded
parents, one legal writer stated, “From the perspective of the
law, the mentally retarded parent is an oxymoron-in-
waiting” (Hayman, 1990, p. 1202). It may be difficult for
decision makers to transcend schematic responses to these
persons.
Three Assumptions Regarding Psychiatrically
Diagnosed Parents
Three assumptions regarding psychiatric diagnosis and
parenting incompetence may unduly bias the evaluation of
parental fitness. It must be emphasized that our aim is not to
dispute that a mental disorder may compromise or disrupt
parenting behaviors but to challenge predisposing biases as
to the level of disruption (e.g., unfitness) and to advocate a
functional–contextual analysis of parenting competencies.
Does a Diagnosis (Past or Present) Predict
Inadequate Parenting or Risk for Child
Maltreatment?
The determination of parenting fitness in termination
cases typically involves two judgments on the part of expert
witnesses: judgments regarding risk of harm to a child if
returned to the parent and judgments regarding whether
caregiving falls below minimally adequate community stan-
dards. Clearly, high proportions of termination cases in-
volve already identified child maltreaters. Although there is
little national data, the information currently available sug-
gests that the majority of these cases involve neglectful
parents rather than abusive ones (Berkowitz & Sedlak,
1993; Jellinek et al., 1992). Despite this, beliefs about
mentally ill individuals and their risk for violence may
influence judgments regarding child risk. Mentally ill per-
sons have been feared by society, in part because of a belief
in their tendency for violence. Whether the mentally ill are
more violent than others has been a subject of considerable
research and debate as has been violence prediction by
mental health professionals (Brody, 1990; Cocozza &
Steadman, 1978; Monahan, 1981; Szmukler, 2001).
Some evidence supports increased risk of violence glo-
bally among the mentally ill. For example, in a large epi-
demiological study of over 7,000 community residents, the
relative risk of aggression by mentally ill persons was
greater than for those without psychiatric illness (two- to
threefold; Swanson, 1994). However, the absolute risk was
not nearly so high. Only 7% of all those with a major mental
disorder (but without substance abuse) engaged in any as-
saultive behavior in a given year. Substance abuse was
associated with a greater risk for violence than mental
illness. A recent study found the prevalence of violence
among mentally ill patients discharged from a hospital to be
the same as the prevalence of violence among controls
living in their communities when neither group had sub-
stance abuse symptoms (Steadman et al., 1998). Violence
was greater among both discharged patients and community
controls with substance abuse compared with those without,
and the greatest violence prevalence was found in
substance-abusing discharged patients.
Although still controversial, some predictive validity for
violence risk has been reported in narrowly defined groups
(e.g., for mental patients with specific diagnoses), under
specific environmental constraints (e.g., in hospital set-
tings), and within short time frames (e.g., immediately
postdischarge; Gardner, Lidz, Mulvey, & Shaw, 1996). Un-
der such circumstances, statements defining level of risk
rather than dichotomous predictions are thought to be pos-
sible (Grisso & Appelbaum, 1992). Most recently the
Macarthur Violence Risk Assessment Study (Monahan et
al., 2000) has developed a tree-based method that is more
accurate than other risk assessment methods and provides a
multiple-risk level assessment rather than a dichotomous
one. However, this is still not widely available for clinical
use nor is it designed to predict child abuse risk per se.
Despite the advances in the actuarial prediction of vio-
lence, when risk prediction regarding child abuse and ne-
glect is considered, available data fail to meet the current
standards for risk prediction outlined by the field (see Grisso
& Appelbaum, 1992). Moreover, it may never be possible to
meet such criteria given the nature of risk involved (within
a particular relationship) and the period over which predic-
tion is required (e.g., for a 2-year-old, it is l6 years until
adulthood). Only a very small literature exists regarding risk
prediction in child maltreatment, and it largely addresses
physical abuse but not neglect. Among already identified
abusive and neglectful populations, follow-up studies with
240BENJET, AZAR, AND KUERSTEN-HOGAN

Page 4

large samples have found a number of predictors of recid-
ivism, including perpetrator characteristics (e.g., the nature
of maltreatment, level of stress, poverty, history of abuse,
expectations of children, social isolation) and child factors
(gender, age; Fryer & Miyoshi, 1994). Mental illness as a
predictor has been found to have low specificity.
Although the question that is before expert witnesses is
one of documented functional significance of the disorder
for the child, strong beliefs within the field may exist that
poor parenting is inherent in diagnosis. Child risk is often
automatically assumed for children of parents diagnosed
with substance abuse, major mental illness (affective and
psychotic disorders), and mental retardation. Each of these
parental conditions is briefly discussed with respect to risk
of harm to the child and minimally adequate parenting.
Substance-abusing parents.
major predictor of termination of parental rights (Jellinek, et
al., 1992). Although there is evidence that such parents are
overrepresented in child protective caseloads (Famularo,
Stone, Barnum, & Wharton, 1986; Murphy et al., 1991), it
is unclear whether this is due to a detection bias, in which
parents are more likely to be identified and reported for
maltreatment because of professionals’ beliefs in greater
child risk, or whether there is an actual increased risk of
child maltreatment. Additionally, children exposed prena-
tally to substances might be more difficult to parent, increas-
ing the likelihood of abusive parental behaviors. Wasser-
man and Leventhal (1993) did find more evidence of both
physical abuse and neglect in medical records of children
born to cocaine-dependent mothers than those born to non-
cocaine-dependent controls, and there are data suggesting
that assaultive behavior (in general, not specific to child
maltreatment) increases in substance-abusing individuals
(Steadman et al., 1998; Swanson, 1994). However, epide-
miological data regarding violence toward children specif-
ically are lacking, and research on the direct links between
substance abuse and inadequate parenting is also scarce.
Maternal substance abuse might compromise the devel-
opment and well-being of a child either by prenatal expo-
sure to drugs and/or alcohol or by impairment of parenting
capabilities postnatally. Evidence for the deleterious effects
of prenatal drug exposure is mixed and varies depending on
the type of substance exposure. For example, initial reports
by the media of the dramatic impairments of cocaine babies
were overemphasized and not confirmed by subsequent
research (Carmichael-Olson & Burgess, 1997). The re-
search findings on the effects of prenatal cocaine exposure
have been inconsistent (Myers, 1992). In contrast, research
on infants exposed to heroin and methadone more consis-
tently points to early problems such as lower birth weights,
higher levels of prematurity, and more respiratory distress
(Householder, Hatcher, Burns, & Chasnoff, 1982). Mayes
(1995) reviews several studies that indicate impairment in
infant state regulation and responsiveness due to fetal alco-
hol effects and neonatal withdrawal syndrome from opiates.
How substance abuse might affect actual parenting be-
havior postnatally is more elusive. Substance abuse has
been thought to interfere negatively with the parent–child
relationship as well as negatively affect the quality of the
environment provided by the substance-abusing parent.
Parental drug abuse is a
Kaplan-Sanoff and Fitzgerald Rice (1992) observed that
parents with substance addiction “have a primary relation-
ship with their drug, not their child” (p.17). Thus, it is often
assumed that substance abuse automatically leads to inade-
quate parenting. Data support the finding of increased be-
havioral difficulties, psychopathology, and substance abuse
among offspring of such parents (Deren, 1986; West &
Prinz, 1987). However, inadequate parenting is not the only
explanation for this finding. For instance, less favorable
child outcome may be due to prenatal drug exposure rather
than an impaired parenting environment. Additionally, there
may be a pre-existing psychiatric or neuropsychological
disorder or a genetic vulnerability that predisposes these
parents to substance abuse and also carries a genetic risk for
their children. The minimal research available on the direct
effects of substance abuse on parenting behaviors or abili-
ties has primarily involved mothers and small samples and
has yielded contradictory results. For example, although one
study found opioid-dependent mothers to be less responsive
and harsher than control mothers (though there was no
difference in guidance and encouragement; Hans, Bernstein,
& Henson, 1999), others have not found greater neglect
for children of substance-abusing mothers (Harrington,
Dubowitz, Black, & Binder, 1995). Heterogeneity of the
substances used and severity as well as the young age of the
children limit the conclusions that can be drawn from these
studies.
Most studies have focused on the two dimensions of
parenting that have emerged repeatedly as important to
child outcome in the developmental literature: maternal
control (also conceived of as restrictiveness, monitoring, or
authoritarianism) and maternal responsiveness (also termed
warmth or involvement; Maccoby & Martin, 1983). Ex-
tremes of control (either being overly permissive or restric-
tive) coupled with low responsiveness have been found to
undermine children’s development. One could imagine that
a parent whose drug of choice is a depressant, might under-
respond to and withdraw from her or his child, thus, inter-
acting with low responsiveness and perhaps providing in-
adequate supervision and neglecting children’s basic needs.
In contrast, a mother addicted to stimulants, might have a
primary mode of interaction with her child that is charac-
terized by reacting unpredictably and impulsively, being
overly responsive, or inconsistently controlling, behaviors
that could possibly escalate to abuse.
Other potential effects of drug use that may bode poorly
for parenting capacities are potential neurological impair-
ments associated with long-term substance use such as
memory and attention, but how or whether such impair-
ments influence parenting has not yet been examined. Also
influenced by substance abuse might be a mother’s affect
and impulse regulation and modulation of anxiety and frus-
tration, both areas in which impairment might lead to in-
creased risk for abuse. Again, well-controlled studies exam-
ining the implications of these issues for parenting are
lacking. (For a thorough review of the literature on sub-
stance abuse and parenting, see Mayes, 1995.)
In addition, the context in which substance-abusing
mothers live might put their children at risk (e.g., high-
crime neighborhood, poverty, marital instability). For ex-
241SPECIAL ISSUE: PARENTAL FITNESS AND MENTAL ILLNESS

Page 5

ample, abuse of illegal drugs may expose a child to in-
creased violence or parental separations on account of
incarceration. Wasserman and Leventhal (1993) found that
cocaine-addicted mothers and their infants had significantly
more separations in the first 2 years of life than a matched
control sample.
Though parental substance abuse most probably carries
risk for impaired parenting, the evidence for specific risk for
any certain type of substance is quite limited, and research
has yet to focus on exactly how parenting behaviors are
affected and whether all parents are similarly affected. The
type of substance chosen and the chronicity of usage may
affect the type and level of risk. In addition, the data
regarding whether parenting risk continues once substance
abuse stops is very limited, but such continued risk may be
of contention in court. The findings of one intervention
study suggest that effects on child outcome that are not
physiologically driven may be found to decrease or disap-
pear with cessation of the substance use (Rodning, Beck-
with, & Howard, 1991).
Staff members in a program offering clinical services for
substance-involved families in the child welfare system
noted that decisions were often based solely on current use
of alcohol and drugs (such as urine screens in a treatment
program; Olsen, Allen, & Azzi-Lessing, 1996). They found
decision making based on such limited criteria inadequate,
as many of the parents who had not made much progress in
substance abuse treatment had adequate parenting skills,
whereas some who appeared to make good progress in
treatment had very limited parenting skills, putting their
children at risk. There is a crucial need for research that
identifies exactly which parenting capacities might be com-
promised and whether they remain compromised if sub-
stance abuse is treated successfully or spontaneous remis-
sion occurs.
Parents with major mental illness.
fering from major psychiatric disorders, such as schizophre-
nia and affective disorders, are a second diagnostic group
whose parenting might come under legal scrutiny. The
affective and cognitive disturbances seen in schizophrenic
parents have been thought to influence parenting capacity
by interfering with parent–child social interactions, whereas
symptoms present in affective disorders such as withdrawal,
irritability, and anhedonia have been thought to influence
the parents’ capacity to be warm and responsive toward
their children and to control their children’s behavior ap-
propriately and consistently. Although such behaviors may
influence parenting, the question remains as to whether they
do in fact constitute unfitness such that parenting is below
acceptable community standards.
Clearly, as with substance abuse, children of parents with
major affective disorders have been shown to have an
increased risk of affective disorder specifically, as well as
other psychological problems more generally, including be-
havior problems, attention deficits, learning disabilities,
cognitive and social deficits, substance abuse, anxiety, and
somatic symptoms (see for example, Beardslee, Versage, &
Gladstone, 1998; Weissman et al., 1987). Similarly, chil-
dren of schizophrenic individuals have been shown to have
an increased risk of psychiatric disorder (Watt, 1984). How-
Those who are suf-
ever, it is not clear what specifically places such children at
risk.
Genetic or biological factors may play some role. For
example, there is considerable evidence for the genetic
transmission of depressive disorders based on twin and
adoption studies. A recent review of twin studies of psychi-
atric illnesses documents in the literature an average heri-
tability of liability to major depression of .33 and a range for
schizophrenia from .60 to .84 (Kendler, 2001). Although the
heightened problems found among offspring are partly due
to genetic forces that have exerted their influence before the
child was even born, we do not know how much is due to
these ill parents’ capacity to be a “good enough” parent and
how much may be due to some third mediating variable that
underlies both the parental illness and the child’s difficul-
ties. In any case, the issue of genetic risk is relevant for
evaluating parents. Children with a genetic vulnerability
probably require more sensitive and optimal parenting than
children without genetic vulnerability and may be prone to
developing problems even when parenting is adequate.
Some research has focused on the parenting capacities of
mentally ill parents. It must be kept in mind that this work
has been designed primarily to evaluate optimal parenting
responses among such parents rather than responses indi-
cating parenting that is below community standards, which
is the threshold set in termination of parental rights cases.
There is some evidence that schizophrenic and depressed
individuals differ in their interactions with their children
compared with nondisordered parents. (Oyserman, Mow-
bray, Meares, & Firminger, 2000, present a review of stud-
ies examining these parenting differences among mothers
with differing diagnoses.) For example, affectively disor-
dered mothers have been found to overreact to mild stres-
sors they experience with their children, such as waiting in
a doctor’s office (Breznitz & Sherman, 1987). Other studies
have found that depressed mothers are often less consistent
toward their children (i.e., ranging from withdrawn to con-
trolling or intrusive) with few mothers falling within the
optimum range of involvement compared with nonde-
pressed mothers (Hoffman & Drotar, 1991). Furthermore,
depressed mothers have been found to use fewer questions
and a less positive tone of voice and to use more criticism
and coercion with their children (Cox, Puckering, Pound, &
Mills, 1987).
It is noteworthy that there is heterogeneity in the quality
of parenting styles of depressed mothers. For example, in
one study, depressed mothers who enjoyed interactions with
their children or received positive responses from their
children were found to be more effective in sustaining
positive interactions with their children (Cox et al., 1987).
This study suggests that the nature of the child’s behavioral
style or temperament must be considered. That is, the
child’s special needs or strengths may exacerbate or temper
the level of risk present. Severity and chronicity of parental
disorder, child’s age at time of onset, child behavior, pa-
rental functioning in the community, and the social support
of the parent also have an impact on child outcome in
families with major disorders (Oyserman, et al., 2000),
although the evidence for some of these risk factors is
inconclusive. Marital discord, single-parent status, social
242BENJET, AZAR, AND KUERSTEN-HOGAN